Automatic Modulation Recognition: A Few-Shot Learning Method Based on the Capsule Network

Li, Lixin; Huang, Junsheng; Cheng, Qianqian; Meng, Hongying; Han, Zhu

doi:10.1109/lwc.2020.3034913

Cited by 54 publications

(22 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Of course, artificial intelligence is not only as simple as the intelligence of traditional equipment, this is only a shallow application of artificial intelligence, and in the context of the continuous development of computer science, artificial intelligence will undoubtedly further develop to the depths of intelligence [ 17 ]. As a branch of computer science, artificial intelligence to research and development of humanized mechanical intelligence (including human action ability, cognitive ability, and thinking ability) as the core, the first through the imitation of humanized intelligence and expansion, lay the basic direction of the development of mechanical intelligence, based on this in-depth research and development, and ultimately achieve the goal of making the degree of mechanical intelligence beyond the human [ 16 , 17 , 21 ]. At present, artificial intelligence technology has been applied to many fields of social production and life, including intelligent robot manufacturing, the development and application of intelligent sensor identification system, digital intelligent payment technology, the development and application of intelligent database, and the development and application of intelligent network security defense system.…”

Section: Related Workmentioning

confidence: 99%

Design of Assessment Judging Model for Physical Education Professional Skills Course Based on Convolutional Neural Network and Few-Shot Learning

Chen

Dong

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

In recent years, the promotion of quality education and the development of curriculum and teaching materials reform have put forward higher goals and requirements for professional skills in physical education. However, there are still many shortcomings in the nuclear assessment of physical education professional skills, such as lack of clarity in evaluation objectives, lack of scientific in evaluation indexes, lack of systematization in evaluation contents, lack of diversity in evaluation methods, lack of authority in evaluation results, and lack of timely prediction and analysis of students’ mastery of classroom teaching skills, thus not giving good play to all the functions that the nuclear assessment should have, thus to a certain extent fettering the further enhancement of physical education. With the development of information technology, artificial intelligence, as a new technology, can guide the improvement of the assessment and judging mode of physical education professional skills courses, and is also an important guiding idea for physical education majors to meet the development demands of information-based society. Based on the analysis of the connotation and characteristics of deep learning, this paper points out the insufficiency of the assessment and evaluation of traditional physical education professional skills courses and proposes a method of assessment and evaluation of physical education professional skills courses based on convolutional neural networks and small sample learning. In the case of a small amount of data in the course assessment, we use a small number of samples to learn, and only need a small number of samples to learn quickly. Using the improvement measures under the teaching concept of deep learning, physical education personnel are required to truly change in terms of professional skills mastery and evaluation. We effectively implement improvement measures, promote the improvement of physical education professional skills, and realize the migration and innovation of sports knowledge and skills.

show abstract

Section: Related Workmentioning

confidence: 99%

Design of Assessment Judging Model for Physical Education Professional Skills Course Based on Convolutional Neural Network and Few-Shot Learning

Chen

Dong

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…This specificity is achieved by means of capsules, which are multiple parallel and independent nodes that learn class specific characteristics and represent points of fixation in the form of a capsule vector. In the context of classifying digitally modulated signals, the capsules are expected to discover excursion characteristics that are intrinsic to a modulation type [7], [14].…”

Section: Capsule Network For Digital Modulationmentioning

confidence: 99%

“…Blind classification of digitally modulated signals has usually been accomplished using signal processing techniques that include likelihood-based approaches [5], [6] or cyclostationary signal processing (CSP) [7]. We note that, unlike the likelihood-or CSP-based approaches, which require the implementation of complex algorithms to extract features that distinguish different digital modulation schemes and then use these features for signal classification, machine learning implements neural networks and relies on their extensive training to make the distinction among different classes of digitally modulated signals.…”

Section: Introductionmentioning

confidence: 99%

“…As discussed in [13], capsule networks are a version of convolutional neural networks that emphasize learning desirable characteristics of the training dataset by means of capsules, which are multiple parallel and independent nodes that learn class specific characteristics. Capsule networks have been used for digital modulation classification in [7], [14], and in our paper we pursue a variant of capsule networks which classifies with high accuracy modulated signals corresponding to several commonly used digital modulation schemes.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Using Capsule Networks to Classify Digitally Modulated Signals with Raw I/Q Data

Latshaw¹,

Popescu²,

Snoap³

et al. 2022

Preprint

View full text Add to dashboard Cite

Machine learning has become a powerful tool for solving problems in various engineering and science areas, including the area of communication systems. This paper presents the use of capsule networks for classification of digitally modulated signals using the I/Q signal components. The generalization ability of a trained capsule network to correctly classify the classes of digitally modulated signals that it has been trained to recognize is also studied by using two different datasets that contain similar classes of digitally modulated signals but that have been generated independently. Results indicate that the capsule networks are able to achieve high classification accuracy. However, these networks are susceptible to the datashift problem which will be discussed in this paper.

show abstract

“…According to the survival/efficiency ratio of UAV considered in literature [10,15,22,23], for environmental perception in extreme situation, route planning is carried out by using the overall survivability and task assignment of the UAV network to guarantee the overall survivability of the UAV network. Another path planning of the UAV network can be summarized as the dynamic UAV network corresponding to the dynamic target group, that is, both the target and the UAV platform have high dynamic capability [14,24]. In this case, the path of the UAV platform is not set specifically, but the rules of its movement are designed to adapt to the changing environment.…”

Section: Inductionmentioning

confidence: 99%

UAV‐Based Collaborative Electronic Reconnaissance Network for 6G

Yang

Song

Xiong

et al. 2021

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

In unmanned aerial vehicle (UAV) collaborative electronic reconnaissance network, single UAV is always restricted by flyability and sensing capacity; hence, a cooperative network is built to realize the electronic reconnaissance. In this paper, a three-level electronic reconnaissance network is proposed, including the radiation target, UAV-based electronic reconnaissance equipment, and the command center. Each of the UAVs is capable of monitoring several radiation targets at the same time. Since the topology of the UAV network influences the effect of electronic reconnaissance, in this contribution, optimization is achieved based on the improvement of radiation coverage. If there is no radiation target within the sensing scope, the corresponding UAV will remove according to our novel strategy. Iterate operations are carried out for the relative optimum performance. Simulation results show that the UAV network topology optimization is capable of improving the coverage of radiation targets effectively.

show abstract

Automatic Modulation Recognition: A Few-Shot Learning Method Based on the Capsule Network

Cited by 54 publications

References 10 publications

Design of Assessment Judging Model for Physical Education Professional Skills Course Based on Convolutional Neural Network and Few-Shot Learning

Design of Assessment Judging Model for Physical Education Professional Skills Course Based on Convolutional Neural Network and Few-Shot Learning

Using Capsule Networks to Classify Digitally Modulated Signals with Raw I/Q Data

UAV‐Based Collaborative Electronic Reconnaissance Network for 6G

Contact Info

Product

Resources

About